'Irish Roots' archive

---

---

Search the entire archive 2009-2016

Search term

Irish Roots

April 20 2015

Transcriptions that don't match

IrishGenealogy.ie has recently re-launched new versions of the civil registration indexes removed with unseemly haste 10 months ago: birth indexes now up to 1913, marriages to 1938 and deaths to 1963.

The good news is that access remains completely free, everything is easily browsable, and entries after 1900 contain significantly more information than the copies already available from the Mormons on FamilySearch.org (and FindMyPast and Ancestry). Birth index entries from 1900 now record the mother's maiden name, making it possible to reconstruct families with some degree of probability, as is already the case on FamilySearch from 1928. And the marriage indexes supply the spouse's name and exact dates starting in 1903, removing uncertainty from existing index searches between 1903 and 1938. Two hurrahs, so far.

However, my initial searches immediately turned up gaps, entries listed in the FamilySearch index that simply don't come up on IrishGenealogy.

Counting gift horses' teeth heads my list of hobbies, so I set about cross-checking record numbers in the two versions.

Looking at the record-counts for all three events in three well-separated registration districts every year between 1880 and 1913 made a number of patterns very clear. First, only 27 out of the 306 counts were identical. Remember, these are supposed to be transcripts of the same records: less than 10 per cent of the tallies actually match. In the vast majority of cases, the Mormon transcript includes more than IrishGenealogy. For births, this came to more than 2000 records, almost 4 per cent of the total over these 34 years. Some of this may be due to flaws in the FamilySearch version - the Mormons regularly transcribe things more than once (they seem to have particularly relished some of the 1880s death indexes, which they transcribed twice, in toto).

But many of these discrepancies are transcription omissions. Which is why having two copies is so important - it's extremely unlikely both sets commit the same errors. And having two sets is what also makes it possible to see the flaws in each. A salutary exercise, if a little hair-raising.

I'll be sweating both sets for all they're worth, but only in full-body sceptic's armour and supping with a long spoon.

[The data used for this piece is at http://johngrenham.com/temporary/GRO%20IrishG.xls. It would be great to see the whole record-set properly mapped like this.]

Blog

2016

Jan	Feb	Mar	Apr	May	June	July	Aug	Sept	Oct	Nov	Dec
4	1
11	8
18
25

2015

Jan	Feb	Mar	Apr	May	June	July	Aug	Sept	Oct	Nov	Dec
5	2	2	6	4	1	6	3	15	5	2	7
13	9	9	13	11	8	13	10	14	12	9	14
19	16	16	20	18	16	20	17	21	19	16	21
26	23	23	29	25	22	27	24	28	26	23	28
		30			29		31			30

2014

Jan	Feb	Mar	Apr	May	June	July	Aug	Sept	Oct	Nov	Dec
6	3	3	7	5	2	7	4	1	6	3	1
13	10	10	14	12	9	14	11	8	13	10	8
20	17	17	21	19	16	21	18	15	20	17	15
27	24	24	28	26	23	28	25	22	27	24	22
		31			30			29			29

---

2013

Jan	Feb	Mar	Apr	May	June	July	Aug	Sept	Oct	Nov	Dec
31	7	4	4	1	6	3	1	5	2	7	4	2
14	11	11	8	13	10	8	12	9	14	11	9
21	18	18	15	20	17	15	19	16	21	18	16
28	25	25	22	27		22	26	23	28	25	23
			29			29		30			30

---

2012

Jan	Feb	Mar	Apr	May	June	July	Aug	Sept	Oct	Nov	Dec
2	6	5	2	7	4	2	6	3	1	5	3
9	13	12	9	14	11	9	13	10	8	12	10
16	20	19	16	21	18	16	20	17	15	19	17
23	27	26	23	28	25	23	27	24	22	26	24
30			30			30			29

---

2011

Jan	Feb	Mar	Apr	May	June	July	Aug	Sept	Oct	Nov	Dec
3	7	7	4	2	6	4	1	5	3	7	5
10	14	14	11	9	13	11	8	12	10	14	12
17	21	21	18	16	20	18	15	19	17	21	19
24	28	28	25	23	27	25	22	26	24	28	26
31				30			29		31

---

2010

Jan	Feb	Mar	Apr	May	June	July	Aug	Sept	Oct	Nov	Dec
4	1	1	5	3	7	5	2	6	4	1	6
11	8	8	12	10	14	12	9	13	11	8	13
18	15	15	19	17	21	19	16	20	18	15	20
25	22	22	26	25	28	26	23	27	25	22	27
				31			30			29

---

2009

Jan	Feb	Mar	Apr	May	June	July	Aug	Sept	Oct	Nov	Dec
6	6	2	2	6	4	1	6	3	7	5	2	7
13	9	9	13	11	8	13	10	14	12	9	14
20	16	16	20	18	15	20	17	21	19	16	21
27	23	23	27	25	22	27	24	28	26	23	28
					29		31			30

---